Augmentation of CD47/SIRPα signaling protects cones in genetic models of retinal degeneration

Abstract

Inherited retinal diseases such as retinitis pigmentosa (RP) can be caused by thousands of different mutations, a small number of which have been successfully treated with gene replacement. However, this approach has yet to scale and may not be feasible in many cases, highlighting the need for interventions that could benefit more patients. Here, we found that microglial phagocytosis is upregulated during cone degeneration in RP, suggesting that expression of "don't eat me" signals such as CD47 might confer protection to cones. To test this, we delivered an adeno-associated viral (AAV) vector expressing CD47 on cones, which promoted cone survival in three mouse models of RP and preserved visual function. Cone rescue with CD47 required a known interacting protein, signal regulatory protein alpha (SIRPα), but not an alternative interacting protein, thrombospondin-1 (TSP1). Despite the correlation between increased microglial phagocytosis and cone death, microglia were dispensable for the prosurvival activity of CD47, suggesting that CD47 interacts with SIRPα on non-microglial cells to alleviate degeneration. These findings establish augmentation of CD47-SIRPα signaling as a potential treatment strategy for RP and possibly other forms of neurodegeneration.

Targeting Microglia to Treat Degenerative Eye Diseases

“…In rd10 mice, which carry a mutation that causes autosomal recessive RP (31), early genetic ablation of microglia using a diphtheria toxin system improved the survival of rod photoreceptors (32). However, in rd1 animals, which harbor a more severe mutation in the same gene, microglia removal using PLX3397 or PLX5622 did not alter the rate of photoreceptor death (20)(21)(22). For RGCs, microglia depletion with PLX5622 had no observable effect after optic nerve crush and failed to preserve visual function in mice injected with microbeads to raise intraocular pressure (IOP) (23,24).…”

“…Since higher doses of radiation eliminate more microglia (39), these data argue that neuroprotection from ionizing radiation is not solely due to microglia depletion. Indeed, while irradiation helped in animal models of RP and glaucoma, treatment with CSF1R inhibitors, which more efficiently remove microglia, did not (20)(21)(22)24). Because microglia depletion after irradiation is relatively slow and incomplete, some of its therapeutic effects might come from changes induced in the microglia that survive.…”

“…Long-term ablation of microglia results in failure to maintain the synapses used by photoreceptors, leading to impaired retinal signaling (41). Furthermore, microglia in some conditions appear to perform favorable actions, since elimination of these cells has a neutral or even detrimental effect on disease progression (19)(20)(21)(22)(23)(24). For these reasons, reprogramming rather than removing microglia may be an appealing therapeutic approach to correct microglial dysregulation while preserving or enhancing their beneficial functions.…”

“…In 2015, Zhao et al reported that microglia in the rd10 model of RP phagocytose living rods, thereby accelerating retinal degeneration (32). Similarly, when incubated with fluorescent particles ex vivo, microglia from rd1 retinas showed much higher phagocytotic activity than those from heterozygous mice (22).…”

“…At the same time, there is a growing appreciation that microglia can also take on beneficial roles during eye disease. In a handful of animal models, improved tools to experimentally remove microglia have unexpectedly revealed these cells to be neutral or even protective against degeneration (19)(20)(21)(22)(23)(24). Microglia may further exhibit distinct functions depending on the stage and chronicity of the disorder, and microglial populations at different physical locations in the eye might behave differently.…”